Introduction to the Theory of HESSocial Dilemmas and Collective Action in Human-Environment Systems Part I - November 3, 2008Prof. Stefanie Engel, Chair of Environmental Policy and EconomicsCHN K 76.3, firstname.lastname@example.org
Outline • Concepts (Social dilemmas, Public Goods, ‚Tragedy of the Commons‘) • Ex. 1: Community-based natural resource management (CBNRM) • Collective Action & CBNRM (Theories, Success factors) Next class: • Collective Action & CBNRM (cont.) • Experimental evidence • Ex. 2: Collective Action & International Commons
Literature Recommended reading to complement lecture: • Ostrom, E., et al. (1999), Revisiting the Commons: Local Lessons, Global Challenges. Science 284:278-282. • Agrawal, A. (2001), Common Property Institutions and Sustainable Governance of Resources. World Development 29(10):1649-1672. Other main references on the theme: • Olson, M. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups Cambridge: Harvard University Press. • Hardin, G. 1968. The Tragedy of the Commons.Science 162(3859):1243-1248. • Ostrom, E. 1990. Governing the Commons. New York: Cambridge University Press. • Baland, J. and J.P. Platteau. 1996. Halting Degradation of Natural Resources: Is There a Role for Rural Communities? Oxford: Clarendon Press.
A Social Dilemma Class Game • Each of you to make a bid between 0 and 10 CHF • Total sum of bids is multiplied by 1.5 and then evenly distributed among all of you (regardless of your bids) next class • Round 1: No communication; Bids and payoffs are kept fully anonymous vis-á-vis other students; Payoffs handed out in close envelope next class • Round 2: With communication; open bidding; payoffs made known next class
A Social Dilemma Class Game (cont.) • If everyone contributed 10, everyone gets 15 ( gain of 5) • In fact, bids of 10 maximize total payoffs for the class • But: If everyone else contributes 10 and individual i contributes 0, i gets to keep his endowment of 10 and in addition gets 1.5/n Σxj, where n=# of students in the class and Σxj is sum of other students‘ contributions • So, for n=10, i would get 10+0.15*90=23.5 • In general: Individual i‘s payoff with bid xi: 10-xi+1.5/n(xi+Σxj) • For n≥2, payoff-maximizing bid is zero • Yet, if everyone choses to bid zero, everybody has payoff 10 no gain
Social dilemmas • Game simulates a social dilemma: Everyone would be better off if everyone contributes to the common good, but if each one takes others‘ contributions as given she can do better by free riding on others‘ efforts, resulting in lower returns for all • Classic example: Hardin‘s (1968) ‚Tragedy of the Commons‘ next slide (overgrazing on common pasture)
“The tragedy of the commons develops in this way. Picture a pasture open to all. (…) As a rational being, each herdsman seeks to maximize his gain. Explicitly or implicitly, more or less consciously, he asks, "What is the utility to me of adding one more animal to my herd?„ This utility has one negative and one positive component. 1) The positive component is a function of the increment of one animal. Since the herdsman receives all the proceeds from the sale of the additional animal, the positive utility is nearly +1. 2) The negative component is a function of the additional overgrazing created by one more animal. Since, however, the effects of overgrazing are shared by all the herdsmen, the negative utility for any particular decisionmaking herdsman is only a fraction of -1. Adding together the component partial utilities, the rational herdsman concludes that the only sensible course for him to pursue is to add another animal to his herd. And another; and another.... But this is the conclusion reached by each and every rational herdsman sharing a commons. Therein is the tragedy. Each man is locked into a system that compels him to increase his herd without limit-in a world that is limited.“ (Hardin, 1968, p. 1244)
Social dilemmas (cont.) • Examples discussed in this class • Local commons (e.g., grazing lands, community-managed forests, irrigation schemes etc.) • International commons (e.g., climate change, ozon layer, ocean fishery) • Do such situations inevitably lead to ‚Tragedy of the Commons‘? • If not, which factors influence results?
Types of Goods with Potential for Social Dilemmas • Common-pool resources: Natural and human-constructed resources in which (i) exclusion of beneficiaries through physical and institutional means is especially costly, and (ii) exploitation by one user reduces resource availability for others • Ex.? Groundwater basins, irrigation systems, fisheries • Public goods: Difficult to exclude others from benefiting and non-rival in consumption (i.e., one person‘s use does not reduce availability to others) • Ex.? Climate, clean air
Social Dilemmas as ‘Prisoner‘ Dilemma‘The Class Game with 2 Students A pair of strategies is a Nash equilibrium if each player’s strategy is optimal, i.e. maximizes her payoffs given the strategy of the other player Student 2 Cooperate (x=10) Defect (x=0) 7.5, 17.5 15, 15 Cooperate Student 1 17.5, 7.5 Defect 10, 10 • (D, D) is the Nash Equilibrium; But (C, C) is the Social Optimum
Social Dilemmas as ‘Prisoner‘ Dilemma ‘Hardin‘s ‘Tragedy of the Commons‘ • 2 herdsmen, who decide how many animals (1 or 2) to let pasture on common land • If both choose 1, each receives payoff of 5 • If both choose 2, animals underfed animals lose economic value payoff is 4 (2 per animal) • If A puts 1 and B puts 2, payoff is 3 per animal Herdsman B No. of animals 2 1 1 3, 6 5, 5 Herdsman A 4, 4 2 6, 3 • Equilibrium: Both put 2 animals
Ex. 1: Community-based natural resource management (CBNRM) • Natural resources in developing countries often used jointly by local communities (groups of users) • Ex: Common grazing lands; Local forests used for extraction of fuelwood and non-timber forest products; Local fisheries • Socio-anthropological field studies show much variation in outcomes (e.g., Ostrom 1990) • CBNRM does not always result in ‘Tragedy of the Commons’; some communities setting up effective system of use and access rules • Hardin’s view overly pessimistic. How can we explain variation?
Ex. 1: CBNRM (cont.) • Moreover: Over past decades trend, mostly in developing countries, to devolve (at least partially) rights and responsibilities over natural resource management from state to local communities and user groups • ‘Devolution’, ‘Decentralization’, ‘Participatory Management’, ‘Co-Management’ • Ex: 50 countries devolving forest management (FAO 1999), ‘Community Irrigation Management’, Community-based Wildlife Management, etc., etc. • Various reasons: state failure in management (lack of monitoring and enforcement capacity; corruption), social justice considerations, declining national budgets, ground-level and donor-driven movements for participatory management approaches • Idea: ‚Sense of ownership‘ leading to incentives for self-regulation & sustainable use • Results also mixed (cases of success and failures). How to explain this?
CBNRM as a Social Dilemma • Behavior of each resource user affects environmental outcome (e.g., forest quality) • Environmental outcome affects all individuals‘ livelihoods • Everyone better off if collective resource use is reduced, but if each user considers only own immediate benefits, not impact on others: ‚Tragedy of the commons‘ possible ENV. SYSTEM HUMAN SYSTEM
Role of collective action (cont.) • Overcoming ‚the tragedy‘ requires Collective Action/Cooperation • Everyone reducing extraction of fuelwood ( Common-pool resource problem, ‚Appropriation problem‘) • Contributing to afforestation, monitoring activities (Public goods problem, ‚Provision problem‘) • Under what conditions can we expect communities to agree on use/management/access rules and to abide by them? ENV. SYSTEM HUMAN SYSTEM
Theories of Collective Action • Olsen‘s ‚Logic of Collective Action‘ (1965) • Group outcome as sum of individual selfish decisions • Although everyone would be better off from cooperation, each individual overexploits/underprovides due to ignoring effects of her actions on others ( Prisoners‘ Dilemma idea, similar to Hardin, 1968) • Large number of studies formalizing the idea & analyzing effects of different user group characteristics on outcomes • Ex: Baland/Platteau 1996, 1997, 1998, 1999, 2002; Bardhan/Dayton-Johnson 2002; Dayton-Johnson/Bardhan 2002; Farrell/Scotchmer 1988; Stern et al. 2002) • See also Hardin (1982) for review of theory • Agrawal (2001): Synthesis of major studies on CBNRM
The role of group size • Olson (1965): Collective action less likely in larger groups • More recent results on Group Size: • Result still holds in some models (e.g., the appropriation model) # of boats in equilibrium Value of fish stock Y=an-bn2 Initial price of a boat p (# of Players) Social opt.
The role of group size (cont.) • Group Size: • But also: Possible economies of scale in transaction costs of monitoring/enforcement at small-medium group sizes • Relationship may be inverted-U shape (Collective action at first increases with group size, then decreases) (Agrawal/Goyal 2001)
The role of heterogeneity Olson (1965): Collective action less likely in more heterogeneous groups More recent results on Heterogeneity: • Effect depends on context and type of heterogeneity. • Socio-cultural heterogeneity tends to reduce collective action • Economic heterogeneity may in some cases enhance it (more below..) • For ex., rich user with large share of benefits may opt to act unilaterally, while poor users facing resource constraints may contribute less
Is ‚Tragedy‘ inevitable? • Simple Prisoner Dilemma framework while useful to explain ‘Tragedy of the Commons’ cannot explain cases of success • Too narrow to describe the full range of real-world Commons situations • Alternative explanations: • Repeated interactions • Alternative payoff/game structures • Social preferences
Repeated Interactions • If Prisoners‘ Dilemma game is repeated an infinite or uncertain number of times, this introduces the possibility of conditional cooperation and punishment (‚tit-for-tat‘) • ‚Folk Theorem‘: Cooperative outcome is one possible equilibrium • Equilibrium is path-dependent, based on presence of trust • Cooperation also possible when actors also interact on other issues • Degree of social interactions & trust as influencing factors
Different Payoff Structures • Idea: Not all Commons situations may exhibit payoff structure of a Prisoner Dilemma • Payoffs may take form of a game in which Collective Action/Cooperation is an equilibrium outcome
Different Payoff StructuresThe Assurance Game Player B Example: Fishing without Dynamite Cooperate (no dyn.) Defect (dynamite) D 0.5, 0 2, 2 Player A 3, 3 C 0, 0.5 • If A cooperates, it is also better for B to cooperate • If only one cooperates both are worse off Cooperation is one possible equilibrium (no incentive to defect once it is implemented, but getting there requires coordination)
Different Payoff StructuresThe Assurance Game (cont.) • Captures widely observed phenomenon that people tend to cooperate when others do (more below) • Role for communication, trust, leadership (acting first) • Norms help to enhance predictability of others’ reactions/behavior
Assurance games with >2 players Can show that in an assurance game with n players, there are usually two possible Nash Equilibria: • Nobody contributes/cooperates • A fraction of players contributes, the others free-ride (Threshold effect) In small groups the fraction contributing may be 100%, but in larger groups it is usually a subset only.
Different Payoff StructuresThe Chicken game Ex: Maintenance of resource crucial to survival Maintained resource yields 10, maintenance costs 4 Player B • 2 equilibria: (C,D) and (D,C), both of which are social optima • Captures idea that consequences of noncooperative behavior are sufficiently gloomy to induce some players to cooperate even when others do not. • Solutions in practice: Unilateral contribution, ‚Claim of first entrant‘, Lottery Defect Cooperate C 8, 8 6, 10 Player A 10, 6 2, 2 D
Different Payoff StructuresHeterogenous Groups Player B (Poor) Player B (Poor) Ex: Poor cannot afford unilateral contribution Defect Defect Cooperate Cooperate C C 15, 4 15, 4 13, 5 13, 5 Player A (Rich) 17, -1 D D 2, 2 Only equilibrium is: Rich contribute, Poor does not (but C,C is social optimum)
Player B (Poor) Cooperate Defect C 15, 4 13, 5 Player A (Rich) D 17, -1 2, 2 Ex: Rich assumes leadership role, putting into effect a system of sanctions punishing non-contributors, incl. himself Different Payoff StructuresHeterogenous Groups and Leadership Without sanctions (as before) Player B (Poor) Cooperate Defect With punishment of 3 if defect when other cooperates; Cooperator gets the fine 16, 2 C 15, 4 Player A (Rich) D 14, 2 2, 2 Both cooperate Rich gains, poor loses
Insights from various game structures • Game structure & outcomes depends on payoffs • Payoffs depend on various factors which may differ across communities and contexts, e.g., • Type of resource • Distribution of benefits and costs from resource extraction • Exit possibilities (alternatives) • Subsistence constraints (e.g. ability to contribute) • etc. • Differences in such factors may explain differences in outcomes • Also: Relevance of communication, trust, norms, leadership